The Legitimacy of Conventional Criminal Law

Our preferences for statutory criminal law should be reexamined.


For many contemporary criminal law theorists, statutory law is supposedly the gold standard. Many believe that statutory legislation is more democratic than having common law crimes. Some maintain that statutory law is a necessary element of the principle of legality or required by political theories of punishment. And compared with common law crimes, many believe that statutory law is superior because it provides ex ante notice of prohibited conduct, whereas the scope of common law crimes is only clarified through ex post decisions.

But our current preference for statutory crimes is a mistake. Statutory crimes are neither inherently more legitimate nor do they necessarily provide better notice. The current view relies on idealized notions of democratic lawmaking.

Start with legitimacy. Majoritarian approval is not necessary to make statutory law in our system. Legislators do not act as perfect agents of their constituents' views. They may get captured by special interests. Or they may decide that it is in their electoral interests to vote with an impassioned minority over a more apathetic majority. Regardless, legislative approval does not inherently imply democratic legitimacy.

Nor is majoritarian support sufficient to make statutory law. Within legislatures, proposed bills face many nonmajoritarian vetogates. Executives can veto bills that enjoy majority support. The federal Senate is malapportioned, and the rules permit the minority to filibuster. Leadership and heads of committees can kill bills that would otherwise pass the legislature. Legislatures have limited legislative time and often full calendars. In many states, the legislatures are part-time, leading to heavily compressed schedules. The result is that many laws with majoritarian support do not get passed, and existing statutes lacking such support do not get amended or repealed.

Because statutory criminal law is difficult to repeal or amend, even with majoritarian support, it faces an analogous objection to the dead-hand objection in constitutional law: from a democratic perspective, why do we care if a legislature passed a crime, particularly if the legislature approved the law at some time in the distant past? Many old laws remain on the books, and some are desuetudinal. But many are actively enforced. The Controlled Substances Act is 40 years old, as is the Gun Control Act. Prosecutions under these laws compose a large fraction of all federal criminal prosecutions. If both laws came up to a vote today, neither law would likely pass in its current form. Why does legislative approval decades ago confer legitimacy today?

Legal conventions serve many functions. One principal function is to correct for defects in the formal legal system, including those related to legitimacy. In the British system, one might ask what gives a (hereditary) monarch the right to wield executive power? (The famous Monty Python skit asking this question is here.) The answer is that, in modern Britain, the monarch does not actually exercise such power. The British constitution vests de facto executive power in a government accountable to the voters.

Conventional criminal law similarly corrects defects with statutory criminal law. My article does not offer a full theory of criminal law, but I assume that, at least in part, what makes an act properly punishable is that the action breaches societal norms. (This is obviously an oversimplification. I do not contend that any conduct that breaches societal norms is inherently criminal conduct. In the article, I qualify this claim substantially.) But societal norms change, and they change faster than legislatures can respond given all their constraints.

Criminal law conventions provide an answer to the problem of criminal law's legitimacy in the face of changing norms. We recognize many old criminal statutes as legitimate criminal law because the community presently recognizes as wrongful the conduct that these statutes prohibit. When societal values evolve, various political and legal checks (which I will detail more in tomorrow's post) constrain prosecutorial discretion around contemporary community norms. These checks effectively result in an unwritten common law that supplements the statutory law and corrects its defects.

This answer is superior to relying on legislative inaction. Some might contend that Congress's decision not to amend or repeal a law implicitly reflects Congress's acknowledgement that the law is legitimate in its current state. But this ignores how difficult legislating in our system is. Inaction does not connote continued approval.

Another objection to unwritten criminal law is that it leaves individuals without notice about what the law is. Statutes, they contend, provide better notice to the community about what conduct is allowed and what is prohibited compared with unwritten law.

I think this objection is mistaken. People learn de facto unwritten criminal law through observation and partaking in shared culture. We know that adultery is not a real crime because no one has ever heard of anyone being arrested for it, despite the fact that infidelity is not uncommon. Drivers learn the real speed limits by observing how family and friends actually drive. The value of written law in providing notice is overstated. Few people read statute books to learn what is lawful (and most who do are probably interested in specific regulatory offenses). Moreover, many statutes are broad, vague, or implicitly rely on unwritten common law concepts (e.g., "malice aforethought," "fraud," or an "attempt").

As I close, let me provide one major qualification to my argument. I do not claim that the existence of criminal law conventions solves all of criminal law's legitimacy problems. As Adrian Vermeule notes, "[c]onventions are equilibria," and sometimes society settles on equilibria that are "normatively abhorrent." Some paradigmatic historical examples are customs of not prosecuting whites who killed African Americans in the South and the underenforcement of domestic violence crimes. Criminal law conventions do not insulate criminal law from societal defects. To the contrary, criminal law conventions rely on indirect means of enforcement (e.g., political pressure), and societal defects may hamper those enforcement mechanisms. A disenfranchised population cannot elect new prosecutors, pressure city councils to curb police abuse, or demand that legislatures rewrite the law.

But while conventional criminal law may not be perfect, it is better and more legitimate than a purely statutory system. The criminal codification project promises more than it can deliver: clear laws prohibiting specific conduct drafted ex ante by democratically-accountable and responsive legislatures. That is not the real world.

Statutory laws are never perfectly drafted. Small errors or changing times can create large overcriminalization problems. Legislatures often lack the time to maintain their criminal codes, and many legislators are incentivized to placate special interests when drafting laws. Modern statutory systems exist in nonideal conditions. These legislative defects are not readily fixable. Rather than wish them away, we would be better advised to continue developing our unwritten customs and traditions to correct the defects in our written law.

NEXT: The Bizarre Effects of Anchoring

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. There’s such a thing as being contrary just to be contrary, and it’s not a good thing.

    Statutory law often falls short of being ideally democratic. Common law is scarcely democratic at all, it would be better termed “judicial law”. And the judiciary are not a representative branch.

    It might be good to introduce a formal desuetude principle into our law, (Though it risks a series of executives effectively repealing popular laws through inaction.) but affirmative non-statutory law genuinely is a bad thing.

    1. This is the biggest failure of the lawyer profession. There are 15 million common law crimes, 3 million being violent, 100 million internet crimes. There are 2 million prosecutions. 20% of the time these prosecutions have the wrong guy. Meanwhile, the lawyer profession is taking our $trillion andreturning little of value for it. They are making their salaries in worthless, make work jobs. They are misusing criminal investigations to harass political opponents.

      The reason for the dropping crime rate is that Chinese fentanyl is killing 83000 criminals a year. As always, the real remedy is technological. So what does the lawyer do? He defunds the police, and now crime is up again, to generate more worthless, make work jobs for lawyers.

      The country was fed up and angry. Judges saved themselves with mandatory sentencing guidelines. They dropped crime 40%. So conservative Scalia leads a charge to end them across the board. Why? Lawyer unemployment soared after their enectment.

      The lawyer profession is just a criminal enterprise, looking out only for itself. Millions of crime victims should answer them by crushing this interal enemy.

    2. I support a desuetude amendment to the constitution. Any rule in any jurisdiction not enforced for 5 years is void. This would be a mass eradication of stupid lawmaking, no one supports, except the lawyer and special interests.

      1. The general problem with a short period for desuetude, is that failing to enforce a popular law is enormously more politically survivable than affirmatively repealing a popular law. But desuetude makes them functionally equivalent.

        Imagine the application of desuetude to immigration law, for instance: Obama could have basically repealed our immigration laws, without any member of Congress having had a say in the matter.

        On the whole I favor the principle, but it needs to be tempered by making the period long enough to bridge a significant number of electoral cycles.

        1. The voter would suffer from that Obama betrayal. They would vote for people to re-enact the immigration laws.

          1. The “bit rate” of voting is too low to accomplish things like that: You’ve got many issues you care about, and only one vote.

            And if voting got us immigration law enforcement, the first two years of the Trump administration would have seen the border wall completed. Inaction is too easy to get away with.

  2. “Few people read statute books to learn what is lawful. . . .”

    Few people slaughter cows and pigs for meat or separate chaff from wheat for bread too.

    That’s because we have specialists who do these things; e.g. lawyers who read statute books.

    1. To whom does the law belong? These cult criminals have converted our property to their unjust enrichment. Any legal utterance that reads above the 6th grade should be void. It violates the Fifth Amendment procedural due process right to notice.

      1. My comment reads at the high school level, and should be legally void.

  3. If Professor Leidor had limited his case to arguing that judges should retain and use common law powers to fill in gaps left by legislatures, and also to do so by recognizing and formalizing customary law rather by simply making up things on their own as they see best, I wouldn’t disagree with him, although I might want to limit the cases where common-law lawmaking occurred more narrowly than Professor Leider might. After all, the common law powers of judges were retained by the Constitution, it’s a perfectly legitimate power, and federal judges have regularly used common-law like powers to fill in gaps in laws like antitrrust and others. And I’d rather have judges base their gap-gilling opinions on common law, which is subject to review and overriding by the legislature, than on the constitution, which isn’t. A judge-made common law subject to review and revision by the elected legislature can be valuable. Judges do have something to say.

    But Professor Leider appears to go far further than just supporting greater use of common law powers subject to legislative oversight and correction. He appears to be arguing for making common law equal to or superior to legislative law. He does this by attacking the foundational legitimacy of republican forms of government, saying that because, in a republican form of government, intermediaries inevitably fail to perfectly reflect the majority opinion of the demos at the moment, as various elites get to argue with and put pressure on representatives and the deliberative system increases this tendency, he argues that this means legislature-made law isn’t any more legitimate in the sense of representing the popular will than judge-made law. We may as well just have laws made by judges.

    I’ll begin by pointing out that this exact argument was made, in the 1930s and 1940s, bu advocates of Hitler and Stalin. As George Orwell pointed out in in a series of essays responding to their arguments and defending representative democracy, however imperfect, over both Naziism and Soviet communism, the advocates were perfectly right that legislative democracy as practiced only very imperfectly reflects the will of the people; elites get to intercede for much the reasons Professor Leider points out. Moreover, civil rights are hardly perfect either. The flaw, as Orwell pointed out, is thinking in absutes, treating differences of degree as having no value, so that if democracy and civil rights aren’t perfect, they might as well not be there at all. If we can’t have perfect liberty, if government can’t perfectly reflect our will, then we might as well have an all-poserful Fuehrer and be done with it.

    The problem here, of course, is that our logic is binary, following the law of the excluded middle, so that if we can negate the opposition even slightly, we have established our case. Reductio ad absurdum arguments work. If something isn’t perfectly true, it’s false plain and simple.

    But in real human life, the middle is hardly ever excluded. Nearly all of life happens somewhere in the middle. In human, a tertium can pretty much be taken as a datum. This means, as one of many consequences, that nearly all human institutions are imperfect. It means that establishing the imperfection of a human institution, as distinct from a mathematical proposition, simply doesn’t prove that it’s worthless. Many vitally important differences are differences of degree, not kind.

    Professor Leider may well have a good argument for the value of common-law rulemaking. Perhaps he might even have a good argument thst it should be regarded as superior to republican legislation. But the argument given here is just not that argument. “Representative democracy imperfect, therefore judges better” is just as flawed an argument as “representative democracy flawed, therefore Hitler better” was in George Orwell’s day.

    1. Pretending that the law of the excluded middle applies to human affairs and making bogus reductio ad absurdim “imperfect, therefore worthless” arguments seems to be the greatest pons assinorum of modern times. Science has come a long way since Aristotle, and logic has had to get a lot subtler in order to support its advances.

      This is particularly relevant to law. The law of the excluded middle dissappears the minute truth has to be established by means of a method of proof based on observation rather than spiritual intuition, because any given method can in some cases simultaneously both fail to observe enough evidence to establish a proposition and fail to observe enough evidence to establish its opposite. And the minute the law of the excluded middle dissappears, reductio ad absurdum arguments fail to work.

      I continue to be astounded to see law professors crafting arguments as if the entire modern world had never happened. Reductio ad absurdum arguments about human affairs continue to be a favored rhetorical tactic.

      1. Part of the problem is demanding formal logic where heuristics are more appropriate. Many formal fallacies are actually good heuristics in real life.

        1. Brett: I would appreciate an example of where a violation of critical thinking helps people.

          1. An example would be, the question of whether to hire a convicted embezzler as your accountant. Formal logic is limited to propositions that are guaranteed to be always true, no exceptions. So, formal logic can not tell you he will rob you blind, he might, he might not. “He embezzled once, he’ll do it again!” is, logically, a fallacy.

            The heuristic doesn’t care that his robbing you isn’t logically guaranteed, that it seems more likely than not is sufficient.

            Use of heuristics isn’t a violation of critical thinking, it’s merely understanding that logical fallacies are confined to formal logic, and decisions must be made without complete certainty.

  4. The courts could account for evolving societal values by declaring selective enforcement to be an “equal protection” violation even when not based on politically sensitive factors. If authorities truly care that pedestrians use the right half of the crosswalk, they must enforce that law uniformly and not just against the targets of an unrelated criminal investigation. Otherwise it becomes unenforceable.

    (Massachusetts model municipal pedestrian control regulations section 4(b): “Pedestrians shall at all times attempt to cross a roadway using the right half of crosswalks.”)

  5. I wouldn’t go so far as to say that statutory law is illegitimate, as this article appears to. Bur yes Judges do often get to fill gaps into statute that are unclear.

    Main thing I want to clarify … this should generally always be done to the benefit of the defendant, not the prosecutor, right? Like the prosecutor can read the law … and people shouldn’t be punished for following the law but not adopting various social cues.

    I think the ideal justice system for me is textualism, but with a strong presumption to make the statutes as narrow as the text allows. That accounts for cases like Yates where the statute yields weird results. So common law is good for that purpose, but again, the object for me should always be to narrow the law, not make it broader. So a mix of textualism, common law constitutionalism, and minimalism

  6. A rather unconventional notion.

    Mr. D.

  7. From Scalia’s dissent from a denial of certiorari, at 555 U.S. 1024, 1027 (2009), in Sorich v. United States:

    Second and relatedly, this Court has long recognized the “basic principle that a criminal statute must give fair warning of the conduct that it makes a crime.” Bouie v. City of Columbia, 378 U. S. 347, 350 (1964). There is a serious argument that §1346 is nothing more than an invitation for federal courts to develop a common-law crime of unethical conduct. But “the notion of a common-law crime is utterly anathema today,” Rogers v. Tennessee, 532 U. S. 451, 476 (2001) (SCALIA, J., dissenting), and for good reason. It is simply not fair to prosecute someone for a crime that has not been defined until the judicial decision that sends him to jail. “How can the public be expected to know what the statute means when the judges and prosecutors themselves do not know, or must make it up as they go along?” Rybicki, supra, at 160 (Jacobs, J., dissenting).

    1. That last sentence reminds me of one of my pet peeves, that appellate decisions needs not be unanimous. If learned and experienced judges can’t agree on what a law means, how the heck can they expect laymen to know? Far as I’m concerned, if interpretation of a law depends on what judges say, the law should be thrown out as unclear, and the legislature can take another crack at writing it better.

    2. This is fairly to deal with. If state courts want to create new common-law crimes, they can simply resolve them prospectively. And if its legitimate for Congress to delegate rule-making powers to administrative agencies applicable to criminal cases, surely it’s legitimate to delegate common-law lawmaking powers to courts, subject to standards and to subsequent revision by Congress.

      Scalia famously preferred bright lines over standards. But he was wrong to overconstitutionalize that preference. The constitution permits standards, indeed, contains some itself. Which tool to use when is a legislative choice.

      My objection is for judges to claim greater authority or legitimacy than legislatures to make law. But the traditional approach in many states whereby courts use common law to fill in gaps, clarify, and occassionally address unforeseen situations, subject to legislative revision and also to a legislative supremacy pre-emption rule whereby courts can neither contradict a statute nor create a common-law scheme out of alignment with it, I don’t see any constitutional problem with a common-law system.

  8. “Why does legislative approval decades ago confer legitimacy today?”

    This question is backwards, let me fix it.

    “What is illegitimate about legislative approval decades ago?”

  9. I think you have some of good job and identifying problems statutory criminal law. I am skeptical of the argument that common law crime would meaningfully address these problems. But, as you have not laid out and explored these arguments yet (which everyone else commenting seems to have ignored for some reason), I will reserve judgment.

  10. The article raises a question without explicitly asking it: Why are criminal statutes so hard to repeal, even with majoritarian support?

    1. In Massachusetts the Catholic Church strongly resisted formal repeal of some unenforced, obsolete laws around 20 years ago. My memory has faded but they could have been abortion or adultery. Repealing unconstitutional but morally sound laws would send the wrong message. The sponsors or legislative leaders chose to let the bill fail rather than narrow it to include only uncontroversial crimes.

  11. Yeah, no.

Please to post comments